NVIDIA Launches RTX PRO Blackwell Workstation GPUs, Bringing Supercomputer-Class Performance to the Desktop

NVIDIA has introduced the RTX PRO Blackwell series, a new generation of workstation GPUs that deliver supercomputer-level performance in a desktop form factor. The Blackwell architecture is designed for professionals in visualization, simulation, and scientific computing, significantly expanding the power and efficiency of local AI workloads.

Supercomputer Performance in a Desktop

The flagship RTX PRO 6000 Blackwell GPU supports up to 96GB of GDDR7 memory per card. Workstations can be configured with up to four GPUs, reaching a total of 384GB of memory. This allows desktops to handle workloads that previously required rack-mounted servers or cloud infrastructure.

Performance scales with the number of GPUs. One GPU provides 4,000 AI TOPS at 600 watts. Two GPUs provide 8,000 AI TOPS at 1,200 watts. Three GPUs provide 12,000 AI TOPS at 1,800 watts. Four GPUs provide 16,000 AI TOPS at 2,400 watts.

For energy-efficient needs, the RTX PRO 6000 Max-Q delivers up to 14,044 AI TOPS with four GPUs while consuming half the power of the standard configuration.

Advanced Design for Professional Workloads

RTX PRO Blackwell GPUs use passive cooling with fan-assisted airflow to manage heat. The platform is built for real-time rendering, complex simulations, scientific visualization, and large-scale data science tasks.

The GPUs support PCIe Gen 5, providing scalable compute capacity according to workstation needs. Local systems equipped with RTX PRO Blackwell GPUs can now run AI workloads such as Llama 3.3-70B, Mixtral 8x7B, and collaborative workflows on the NVIDIA Omniverse platform with up to three times the productivity of previous generations.

This allows desktop workstations to handle demanding AI workloads that were previously limited to cloud environments, expanding applications in retail, manufacturing, and industrial edge computing.

Integrated AI Ecosystem

The Blackwell generation works with the NVIDIA CUDA-X library stack, Omniverse, AI Enterprise, Blueprints, and NIM microservices. This enables developers to build, optimize, and deploy AI applications across desktops, data centers, and cloud environments.

Workstations equipped with RTX PRO GPUs support local prototyping, accelerated AI development, and enterprise-grade inference services. They can also scale into larger data center deployments when needed.

One Unified Architecture

The RTX PRO Blackwell series is part of NVIDIA’s unified architecture strategy spanning Cloud AI, Enterprise AI, Personal AI, and Edge AI. By delivering high performance, efficiency, and scalability, the GPUs unlock opportunities in generative AI, simulation, scientific research, and content creation. They establish a new standard for desktop professional computing.